- Add eval_concurrency config field with asyncio.Semaphore - Add local.yaml config using Docker backend (sandboxed, no cloud costs) - Register docker_image alongside modal_image for backend flexibility - Default: 8 parallel tasks for local runs |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| default.yaml | ||
| run_eval.sh | ||
| terminalbench2_env.py | ||